DCGAN论文导读、关键点说明及代码实现修改（2）

源码： https://github.com/eriklindernoren/Keras-GAN/blob/master/dcgan/dcgan.py
这个例子中选用的数据集是mnist，代码解析及修改成自己的数据训练使用方法如下：

from __future__ import print_function, division

from keras.datasets import mnist
from keras.layers import Input, Dense, Reshape, Flatten, Dropout
from keras.layers import BatchNormalization, Activation, ZeroPadding2D
from keras.layers.advanced_activations import LeakyReLU
from keras.layers.convolutional import UpSampling2D, Conv2D
from keras.models import Sequential, Model
from keras.optimizers import Adam

import matplotlib.pyplot as plt

import sys

import numpy as np

class DCGAN():  #定义DCGAN的类
    def __init__(self):
        # Input shape
        self.img_rows = 28 
        self.img_cols = 28
        self.channels = 1
        #此处mnist图像就是28x28单通道（灰度图），如果是自己的数据需要修改
        self.img_shape = (self.img_rows, self.img_cols, self.channels)
        self.latent_dim = 100#生成器输入的噪声信号的维度

        optimizer = Adam(0.0002, 0.5)#论文中提到的adam优化器参数设定

        # Build and compile the discriminator
        self.discriminator = self.build_discriminator()
        self.discriminator.compile(loss='binary_crossentropy',
            optimizer=optimizer,
            metrics=['accuracy'])

        # Build the generator
        self.generator = self.build_generator()

        # The generator takes noise as input and generates imgs
        z = Input(shape=(self.latent_dim,))#此处z是生成器的输出
        img = self.generator(z) #img为生成器的输出，可以理解成生成的欺骗判别器的图像

        # For the combined model we will only train the generator
        self.discriminator.trainable = False #当生成器工作的时候，判别器要固定，即不可训练

        # The discriminator takes generated images as input and determines validity
        valid = self.discriminator(img)#判别器为生成器的结果打分

        # The combined model  (stacked generator and discriminator)
        # Trains the generator to fool the discriminator
        self.combined = Model(z, valid)
        self.combined.compile(loss='binary_crossentropy', optimizer=optimizer)

    def build_generator(self):#定义生成器方法

        model = Sequential()
        #如果想修改生成器主要要注意各个层之间数据的shape匹配问题
        model.add(Dense(128 * 7 * 7, activation="relu", input_dim=self.latent_dim))
        model.add(Reshape((7, 7, 128)))
        model.add(UpSampling2D())
        model.add(Conv2D(128, kernel_size=3, padding="same"))
        model.add(BatchNormalization(momentum=0.8))
        model.add(Activation("relu"))
        model.add(UpSampling2D())
        model.add(Conv2D(64, kernel_size=3, padding="same"))
        model.add(BatchNormalization(momentum=0.8))
        model.add(Activation("relu"))
        model.add(Conv2D(self.channels, kernel_size=3, padding="same"))
        model.add(Activation("tanh"))

        model.summary()

        noise = Input(shape=(self.latent_dim,))
        img = model(noise)

        return Model(noise, img)

    def build_discriminator(self):#定义判别器方法

        model = Sequential()

        model.add(Conv2D(32, kernel_size=3, strides=2, input_shape=self.img_shape, padding="same"))
        model.add(LeakyReLU(alpha=0.2))
        model.add(Dropout(0.25))
        model.add(Conv2D(64, kernel_size=3, strides=2, padding="same"))
        model.add(ZeroPadding2D(padding=((0,1),(0,1))))
        model.add(BatchNormalization(momentum=0.8))
        model.add(LeakyReLU(alpha=0.2))
        model.add(Dropout(0.25))
        model.add(Conv2D(128, kernel_size=3, strides=2, padding="same"))
        model.add(BatchNormalization(momentum=0.8))
        model.add(LeakyReLU(alpha=0.2))
        model.add(Dropout(0.25))
        model.add(Conv2D(256, kernel_size=3, strides=1, padding="same"))
        model.add(BatchNormalization(momentum=0.8))
        model.add(LeakyReLU(alpha=0.2))
        model.add(Dropout(0.25))
        model.add(Flatten())
        model.add(Dense(1, activation='sigmoid'))

        model.summary()

        img = Input(shape=self.img_shape)
        validity = model(img)

        return Model(img, validity)

    def train(self, epochs, batch_size=128, save_interval=50):

        # Load the dataset
        (X_train, _), (_, _) = mnist.load_data() #加载训练数据集，后面会说明如何加载自己的数据集

        # Rescale -1 to 1
        X_train = X_train / 127.5 - 1. #输入数据归一化（使模型易于收敛，该区间内sigmoid函数的梯度较大）
        X_train = np.expand_dims(X_train, axis=3) #常规数据shape变换，为了满足模型input需求

        # Adversarial ground truths
        valid = np.ones((batch_size, 1)) #如果数据是mnist中加载的则标签为1
        fake = np.zeros((batch_size, 1)) #如果数据是生成器生成的则标签为0
                                         #此处主要训练判别器时使用

        for epoch in range(epochs):

            # ---------------------
            #  Train Discriminator
            # ---------------------

            # Select a random half of images
            idx = np.random.randint(0, X_train.shape[0], batch_size)
            imgs = X_train[idx]#从训练集中随机拿出batch_size个数据组成一个张量送入网络

            # Sample noise and generate a batch of new images
            noise = np.random.normal(0, 1, (batch_size, self.latent_dim))
            gen_imgs = self.generator.predict(noise)

            # Train the discriminator (real classified as ones and generated as zeros)
            #定义判别器的损失，两部分组成，分别是
            #loss_real和loss_fake，定义如下，权重55开计算总体           
            d_loss_real = self.discriminator.train_on_batch(imgs, valid)
            d_loss_fake = self.discriminator.train_on_batch(gen_imgs, fake)
            d_loss = 0.5 * np.add(d_loss_real, d_loss_fake)

            # ---------------------
            #  Train Generator
            # ---------------------

            # Train the generator (wants discriminator to mistake images as real)
            g_loss = self.combined.train_on_batch(noise, valid)

            # Plot the progress
            print ("%d [D loss: %f, acc.: %.2f%%] [G loss: %f]" % (epoch, d_loss[0], 100*d_loss[1], g_loss))

            # If at save interval => save generated image samples
            if epoch % save_interval == 0: #设定模型保存间隔
                self.save_imgs(epoch)

    def save_imgs(self, epoch):
        r, c = 5, 5  #指定plt中画布中放置多少结果图像
        noise = np.random.normal(0, 1, (r * c, self.latent_dim))
        gen_imgs = self.generator.predict(noise)

        # Rescale images 0 - 1
        gen_imgs = 0.5 * gen_imgs + 0.5

        fig, axs = plt.subplots(r, c)
        cnt = 0
        for i in range(r):
            for j in range(c):
                axs[i,j].imshow(gen_imgs[cnt, :,:,0], cmap='gray')
                axs[i,j].axis('off')
                cnt += 1
        fig.savefig("images/mnist_%d.png" % epoch)
        plt.close()


if __name__ == '__main__':
    dcgan = DCGAN() #类的实例化
    dcgan.train(epochs=4000, batch_size=32, save_interval=50)
     #调用train方法，此处参数 epochs指定总训练次数
     #                     batch_size 的设置参考上一篇文章，建议不小于32