Using Generative Adversarial Network to generate Single Image

A generative adversarial network is comprised of two parts which are necessary for the training, the generator and the discriminator. They are playing a game against each other, the discriminator tried to learn what is real and what is fake, and the generator tries to create fake instances which will fool the discriminator.

At the input of the discriminator you will use real images, for example the MNIST numbers. At the input of the generator you will put just random noise vectors. These vectors will then be shaped by the layers of the generator to create some image, hopefully resembling the MNIST dataset. If it fools the discriminator, it wins that round. Otherwise, it will tune its weights using backpropagation in order to produce better quality results in the next round.

Once training is complete, you will detach the discriminator and use only the generator. Then for every single random noise vector fed to the input of the generator you will receive an associated image.

Here is a GaN implementation. You will need to put a path "./gan/images/" i the directory where you are running this GaN.

import numpy as np import time from tensorflow.examples.tutorials.mnist import input_data from keras.layers import Input from keras.models import Model from keras.models import Sequential from keras.layers import Dense, Dropout, Flatten from keras.layers import Conv2D, MaxPooling2D from keras.layers import Dense, Activation, Flatten, Reshape from keras.layers import Conv2D, Conv2DTranspose, UpSampling2D from keras.layers import LeakyReLU, Dropout from keras.layers import BatchNormalization from keras.optimizers import Adam, RMSprop from keras.datasets import mnist import matplotlib.pyplot as plt class GAN(): def __init__(self): self.img_rows = 28 self.img_cols = 28 self.channels = 1 self.img_shape = (self.img_rows, self.img_cols, self.channels) optimizer = Adam(0.0002, 0.5) # Build and compile the discriminator self.discriminator = self.build_discriminator() self.discriminator.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=['accuracy']) # Build and compile the generator self.generator = self.build_generator() self.generator.compile(loss='binary_crossentropy', optimizer=optimizer) # The generator takes noise as input and generated imgs z = Input(shape=(28,28,)) img = self.generator(z) # For the combined model we will only train the generator self.discriminator.trainable = False # The valid takes generated images as input and determines validity valid = self.discriminator(img) # The combined model (stacked generator and discriminator) takes # noise as input => generates images => determines validity self.combined = Model(z, valid) self.combined.compile(loss='binary_crossentropy', optimizer=optimizer) def build_generator(self): noise_shape = (28,28,) model = Sequential() model.add(Reshape((28*28,), input_shape=noise_shape)) model.add(Dense(256)) model.add(LeakyReLU(alpha=0.2)) model.add(BatchNormalization(momentum=0.8)) model.add(Dense(512)) model.add(LeakyReLU(alpha=0.2)) model.add(BatchNormalization(momentum=0.8)) model.add(Dense(1024)) model.add(LeakyReLU(alpha=0.2)) model.add(BatchNormalization(momentum=0.8)) model.add(Dense(np.prod(self.img_shape), activation='tanh')) model.add(Reshape(self.img_shape)) model.summary() noise = Input(shape=noise_shape) img = model(noise) return Model(noise, img) def build_discriminator(self): img_shape = (self.img_rows, self.img_cols, self.channels) model = Sequential() model.add(Conv2D(32, kernel_size=(3, 3), input_shape=img_shape)) model.add(MaxPooling2D(pool_size=(2, 2))) model.add(LeakyReLU(alpha=0.2)) model.add(Dropout(0.25)) model.add(Flatten()) model.add(Dense(64)) model.add(LeakyReLU(alpha=0.2)) model.add(Dense(1, activation='sigmoid')) model.summary() img = Input(shape=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() # Rescale -1 to 1 X_train = (X_train.astype(np.float32) - 127.5) / 127.5 X_train = np.expand_dims(X_train, axis=3) half_batch = int(batch_size / 2) for epoch in range(epochs): # --------------------- # Train Discriminator # --------------------- # Select a random half batch of images idx = np.random.randint(0, X_train.shape[0], half_batch) imgs = X_train[idx] noise = np.random.normal(0, 1, (half_batch,28,28,)) # Generate a half batch of new images gen_imgs = self.generator.predict(noise) # Train the discriminator d_loss_real = self.discriminator.train_on_batch(imgs, np.ones((half_batch, 1))) d_loss_fake = self.discriminator.train_on_batch(gen_imgs, np.zeros((half_batch, 1))) d_loss = 0.5 * np.add(d_loss_real, d_loss_fake) # --------------------- # Train Generator # --------------------- noise = np.random.normal(0, 1, (batch_size,28,28,)) # The generator wants the discriminator to label the generated samples # as valid (ones) valid_y = np.array([1] * batch_size) # Train the generator g_loss = self.combined.train_on_batch(noise, valid_y) # 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: self.save_imgs(epoch) def save_imgs(self, epoch): r, c = 5, 5 noise = np.random.normal(0, 1, (r * c, 28,28,)) 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("gan/images/mnist_%d.png" % epoch) plt.close() if __name__ == '__main__': gan = GAN() gan.train(epochs=30000, batch_size=64, save_interval=200)