【深度學習】--GAN從入門到初始

一、前述

GAN，生成對抗網絡，在2016年基本火爆深度學習，所有有必要學習一下。生成對抗網絡直觀的應用可以幫我們生成數據，圖片。

二、具體

1、生活案例

比如假設真錢 r  

壞人定義為G  我們通過 G 給定一個噪音X 通過學習一組參數w 生成一個G（x），轉換成一個真實的分布。 這就是生成，相當於造假錢。

警察定義為D 將G(x)和真錢r 分別輸入給判別網絡，能判別出真假，真錢判別為0，假錢判別為1 。這就是判別。

最后生成網絡想讓判別網絡判別不出來什么是真實的，什么是假的。要想生成的更好，則判別的就必須更強。有些博弈的思想，只有你強了，我才更強！！。

2、數學案例

我們最后的希望。

 

 3、損失函數

4、代碼案例

 流程：

為了使判別模型更好，所以我們額外訓練一個D_pre網絡，使得判別模型能夠判別出哪些是0，哪些是1，訓練完之后會得到一組w,b參數。這樣我們在真正初始化判別模型D的時候就能根據之前的D_pre來進行初始化。

 代碼：

import argparse import numpy as np from scipy.stats import norm import tensorflow as tf import matplotlib.pyplot as plt from matplotlib import animation import seaborn as sns sns.set(color_codes=True) seed = 42 np.random.seed(seed) tf.set_random_seed(seed) class DataDistribution(object): def __init__(self): self.mu = 4#均值
        self.sigma = 0.5#標准差

    def sample(self, N): samples = np.random.normal(self.mu, self.sigma, N) samples.sort() return samples class GeneratorDistribution(object):#在生成模型額噪音點，初始化輸入
    def __init__(self, range): self.range = range def sample(self, N): return np.linspace(-self.range, self.range, N) + \ np.random.random(N) * 0.01


def linear(input, output_dim, scope=None, stddev=1.0): norm = tf.random_normal_initializer(stddev=stddev) const = tf.constant_initializer(0.0) with tf.variable_scope(scope or 'linear'): w = tf.get_variable('w', [input.get_shape()[1], output_dim], initializer=norm) b = tf.get_variable('b', [output_dim], initializer=const) return tf.matmul(input, w) + b def generator(input, h_dim): h0 = tf.nn.softplus(linear(input, h_dim, 'g0'))#12*1
    h1 = linear(h0, 1, 'g1') return h1#z最后的生成模型


def discriminator(input, h_dim): h0 = tf.tanh(linear(input, h_dim * 2, 'd0'))#linear 控制初始化參數
    h1 = tf.tanh(linear(h0, h_dim * 2, 'd1')) h2 = tf.tanh(linear(h1, h_dim * 2, scope='d2')) h3 = tf.sigmoid(linear(h2, 1, scope='d3'))#最終的輸出值 對判別網絡輸出
    return h3 def optimizer(loss, var_list, initial_learning_rate): decay = 0.95 num_decay_steps = 150#沒迭代150次 學習率衰減一次0.95-150*0.95
    batch = tf.Variable(0) learning_rate = tf.train.exponential_decay( initial_learning_rate, batch, num_decay_steps, decay, staircase=True ) optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize( loss, global_step=batch, var_list=var_list ) return optimizer class GAN(object): def __init__(self, data, gen, num_steps, batch_size, log_every): self.data = data self.gen = gen self.num_steps = num_steps self.batch_size = batch_size self.log_every = log_every self.mlp_hidden_size = 4#隱層神經元個數
 self.learning_rate = 0.03#學習率
 self._create_model() def _create_model(self): with tf.variable_scope('D_pre'):#構造D_pre模型骨架,預先訓練,為了去初始化真正的判別模型
            self.pre_input = tf.placeholder(tf.float32, shape=(self.batch_size, 1)) self.pre_labels = tf.placeholder(tf.float32, shape=(self.batch_size, 1)) D_pre = discriminator(self.pre_input, self.mlp_hidden_size) self.pre_loss = tf.reduce_mean(tf.square(D_pre - self.pre_labels)) self.pre_opt = optimizer(self.pre_loss, None, self.learning_rate) # This defines the generator network - it takes samples from a noise
        # distribution as input, and passes them through an MLP.
        with tf.variable_scope('Gen'):#生成模型
            self.z = tf.placeholder(tf.float32, shape=(self.batch_size, 1))#噪音的輸入
            self.G = generator(self.z, self.mlp_hidden_size)#最后的生成結果

        # The discriminator tries to tell the difference between samples from the
        # true data distribution (self.x) and the generated samples (self.z).
        #         # Here we create two copies of the discriminator network (that share parameters),
        # as you cannot use the same network with different inputs in TensorFlow.
        with tf.variable_scope('Disc') as scope:#判別模型 不光接受真實的數據 還要接受生成模型的判別
            self.x = tf.placeholder(tf.float32, shape=(self.batch_size, 1)) self.D1 = discriminator(self.x, self.mlp_hidden_size)#真實的數據
            scope.reuse_variables()#變量重用
            self.D2 = discriminator(self.G, self.mlp_hidden_size)#生成的數據

        # Define the loss for discriminator and generator networks (see the original
        # paper for details), and create optimizers for both
        self.loss_d = tf.reduce_mean(-tf.log(self.D1) - tf.log(1 - self.D2))#判別網絡的損失函數
        self.loss_g = tf.reduce_mean(-tf.log(self.D2))#生成網絡的損失函數,希望其趨向於1
 self.d_pre_params = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope='D_pre') self.d_params = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope='Disc') self.g_params = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope='Gen') self.opt_d = optimizer(self.loss_d, self.d_params, self.learning_rate) self.opt_g = optimizer(self.loss_g, self.g_params, self.learning_rate) def train(self): with tf.Session() as session: tf.global_variables_initializer().run() # pretraining discriminator
            num_pretrain_steps = 1000#迭代次數，先訓練D_pre ,先讓其有一個比較好的初始化參數
            for step in range(num_pretrain_steps): d = (np.random.random(self.batch_size) - 0.5) * 10.0 labels = norm.pdf(d, loc=self.data.mu, scale=self.data.sigma) pretrain_loss, _ = session.run([self.pre_loss, self.pre_opt], {#相當於一次迭代
                    self.pre_input: np.reshape(d, (self.batch_size, 1)), self.pre_labels: np.reshape(labels, (self.batch_size, 1)) }) self.weightsD = session.run(self.d_pre_params)#相當於拿到之前的參數
            # copy weights from pre-training over to new D network
            for i, v in enumerate(self.d_params): session.run(v.assign(self.weightsD[i]))#吧權重參數拷貝

            for step in range(self.num_steps):#訓練真正的生成對抗網絡
                # update discriminator
                x = self.data.sample(self.batch_size)#真實的數據
                z = self.gen.sample(self.batch_size)#隨意的數據，噪音點
                loss_d, _ = session.run([self.loss_d, self.opt_d], {#D兩種輸入真實，和生成的
                    self.x: np.reshape(x, (self.batch_size, 1)), self.z: np.reshape(z, (self.batch_size, 1)) }) # update generator
                z = self.gen.sample(self.batch_size)#G網絡
                loss_g, _ = session.run([self.loss_g, self.opt_g], { self.z: np.reshape(z, (self.batch_size, 1)) }) if step % self.log_every == 0: print('{}: {}\t{}'.format(step, loss_d, loss_g)) if step % 100 == 0 or step==0 or step == self.num_steps -1 : self._plot_distributions(session) def _samples(self, session, num_points=10000, num_bins=100): xs = np.linspace(-self.gen.range, self.gen.range, num_points) bins = np.linspace(-self.gen.range, self.gen.range, num_bins) # data distribution
        d = self.data.sample(num_points) pd, _ = np.histogram(d, bins=bins, density=True) # generated samples
        zs = np.linspace(-self.gen.range, self.gen.range, num_points) g = np.zeros((num_points, 1)) for i in range(num_points // self.batch_size): g[self.batch_size * i:self.batch_size * (i + 1)] = session.run(self.G, { self.z: np.reshape( zs[self.batch_size * i:self.batch_size * (i + 1)], (self.batch_size, 1) ) }) pg, _ = np.histogram(g, bins=bins, density=True) return pd, pg def _plot_distributions(self, session): pd, pg = self._samples(session) p_x = np.linspace(-self.gen.range, self.gen.range, len(pd)) f, ax = plt.subplots(1) ax.set_ylim(0, 1) plt.plot(p_x, pd, label='real data') plt.plot(p_x, pg, label='generated data') plt.title('1D Generative Adversarial Network') plt.xlabel('Data values') plt.ylabel('Probability density') plt.legend() plt.show() def main(args): model = GAN( DataDistribution(), GeneratorDistribution(range=8), args.num_steps, args.batch_size, args.log_every, ) model.train() def parse_args(): parser = argparse.ArgumentParser() parser.add_argument('--num-steps', type=int, default=1200, help='the number of training steps to take') parser.add_argument('--batch-size', type=int, default=12, help='the batch size') parser.add_argument('--log-every', type=int, default=10, help='print loss after this many steps') return parser.parse_args() if __name__ == '__main__': main(parse_args())

 

 結果：

迭代到最后時候可以看到結果越來越類似。