def main():
seed = 42
np.random.seed(seed)
current_dir = os.path.dirname(__file__)
sys.path.append(os.path.join(current_dir, '..'))
current_dir = current_dir if current_dir is not '' else '.'
img_dir_path = 'jpg'
txt_dir_path = 'flowers/text_c10'
model_dir_path = current_dir + '/models'
img_width = 32
img_height = 32
img_channels = 3
from dcgan import DCGan
from img_cap_loader import load_normalized_img_and_its_text
image_label_pairs = load_normalized_img_and_its_text(img_dir_path,
txt_dir_path,
img_width=img_width,
img_height=img_height)
shuffle(image_label_pairs)
gan = DCGan()
gan.img_width = img_width
gan.img_height = img_height
gan.img_channels = img_channels
gan.random_input_dim = 200
gan.glove_source_dir_path = './very_large_data'
batch_size = 16
epochs = 300
gan.fit(model_dir_path=model_dir_path,
image_label_pairs=image_label_pairs,
snapshot_dir_path=current_dir + '/data/snapshots',
snapshot_interval=100,
batch_size=batch_size,
epochs=epochs)
gan.img_width = img_width
gan.img_height = img_height
gan.img_channels = img_channels
gan.random_input_dim = 200
gan.glove_source_dir_path = './very_large_data'
batch_size = 5
epochs = 2000
if mode == 'train':
#training
start_time = time.time()
logs = gan.fit(model_dir_path=model_dir_path,
image_label_pairs=image_label_pairs,
snapshot_dir_path=current_dir + snapshots_dir_path,
snapshot_interval=10,
batch_size=batch_size,
epochs=epochs)
from training_plot import on_epoch_end
on_epoch_end(logs)
end_time = time.time()
print(end_time - start_time)
elif mode == 'test':
#testing
from dcgan import DCGan
gan = DCGan()
gan.load_model(model_dir_path)