wgan_gp

wgan_gp_0.py

# celebA dataset path
dataset_path = "Dataset/img_align_celeba"

# Set the input shape and size for the generator and discriminator
batch_size = 128
img_shape = (64, 64, 3) # The shape of the input image, input to the discriminator
noise_dim = 128 # The dimension of the noise vector, input to the generator
model_path = 'Models/02_WGANGP_faces'
os.makedirs(model_path, exist_ok=True)

# Define your data generator
datagen = ImageDataGenerator(
    preprocessing_function=lambda x: (x / 127.5) - 1.0,  # Normalize image pixel values to [-1, 1]
    horizontal_flip=True  # Data augmentation
)

# Create a generator that yields batches of images
train_generator = datagen.flow_from_directory(
    directory=dataset_path,  # Path to directory containing images
    target_size=img_shape[:2],  # Size of images (height, width)
    batch_size=batch_size,
    class_mode=None,  # Do not use labels
    shuffle=True,  # Shuffle the data
)