pdfcoffee

Generative Adversarial Networks
The complete GAN is made by combining the two:
class DCGAN():
def __init__(self, rows, cols, channels, z = 100):
# Input shape
self.img_rows = rows
self.img_cols = cols
self.channels = channels
self.img_shape = (self.img_rows, self.img_cols, self.channels)
self.latent_dim = z
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 the generator
self.generator = self.build_generator()
# The generator takes noise as input and generates imgs
z = Input(shape=(self.latent_dim,))
img = self.generator(z)
# For the combined model we will only train the generator
self.discriminator.trainable = False
# The discriminator takes generated images as input and
# determines validity
valid = self.discriminator(img)
# 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)
The GAN is trained in the same manner as before; first random noise is fed to the
generator. The output of the generator is added with real images to initially train the
discriminator, and then the generator is trained to give an image that can fool the
discriminator. The process is repeated for the next batch of images. The GAN takes
between a few hundred to thousands of epochs to train:
[ 204 ]