def worker_estimator(args, manager, config, make_env):
init_logging_handler(args.log_dir, '_estimator')
agent = GAN(make_env, args, manager, config, args.process, pre_irl=True)
agent.load(args.save_dir + '/best')
best0, best1 = float('inf'), float('inf')
for e in range(args.epoch):
agent.train_disc(e, args.batchsz_traj)
best0 = agent.test_disc(e, args.batchsz, best0)
def main(argv):
# Load configs from file
config = json.load(open(FLAGS.config))
# set_backend()
# Set name
#name = '{}_{}_'.format(config['INPUT_NAME'], config['TARGET_NAME'])
#for l in config['LABELS']:
# name += str(l)
#config['NAME'] += '_' + name
if FLAGS.use_wandb:
import wandb
resume_wandb = True if FLAGS.wandb_resume_id is not None else False
wandb.init(config=config,
resume=resume_wandb,
id=FLAGS.wandb_resume_id,
project='EchoGen',
name=FLAGS.wandb_run_name)
# Initialize GAN
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
model = GAN(config, FLAGS.use_wandb, device, FLAGS.dataset_path)
# load trained models if they exist
if FLAGS.ckpt_load is not None:
model.load(f'{FLAGS.ckpt_load}/generator_last_checkpoint.bin',
model='generator')
model.load(f'{FLAGS.ckpt_load}/discriminator_last_checkpoint.bin',
model='discriminator')
if FLAGS.test:
model.test()
else:
model.train()
generator.add(BatchNormalization(128))
generator.add(Activation(config.nonlinearity))
generator.add(
Deconvolution2D(128,
3,
ksize=4,
stride=2,
pad=paddings.pop(0),
use_weightnorm=config.use_weightnorm))
if config.distribution_output == "sigmoid":
generator.add(sigmoid())
if config.distribution_output == "tanh":
generator.add(tanh())
params = {
"config": config.to_dict(),
"model": generator.to_dict(),
}
with open(generator_sequence_filename, "w") as f:
json.dump(params, f, indent=4, sort_keys=True, separators=(',', ': '))
generator_params = params
gan = GAN(discriminator_params, generator_params)
gan.load(args.model_dir)
if args.gpu_device != -1:
cuda.get_device(args.gpu_device).use()
gan.to_gpu()
random.seed(rnd_seed)
numpy.random.seed(rnd_seed)
torch.manual_seed(rnd_seed)
print("Random Seed: ", rnd_seed)
#CUDA
if torch.cuda.is_available():
Tensor = torch.cuda.FloatTensor
print("CUDA is available")
else:
Tensor = torch.Tensor
print("CUDA is NOT available")
#load model
gan = GAN(opts)
gan.load(opts.output_path)
#display model info
summary(gan.generator.cuda(), input_size=(opts.latent_dim, 1, 1))
summary(gan.discriminator.cuda(),
input_size=(opts.channels_nbr, opts.image_size, opts.image_size))
#IMAGE GENERATION
#test model
nrows = 16
ncols = 16
latent_vectors = Tensor(
numpy.random.normal(0.0, 1.0, (nrows * ncols, opts.latent_dim, 1, 1)))
gen = gan.generator(latent_vectors).detach()
for p in processes:
p.join()
worker_estimator(args, manager, config, make_env)
logging.info("pretrain ends here")
elif args.test:
logging.debug('test')
logging.disable(logging.DEBUG)
current_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime())
# dir_name = datetime.now().isoformat()
args.save_dir = os.path.join('model_' + args.gan_type, current_time)
logging.info(args.save_dir)
agent = GAN(make_env, args, manager, config, 1, infer=True)
agent.load(args.load)
logging.info("model loading finish and start evaluating")
agent.evaluate(save_dialog=True)
# agent.evaluate()
# agent.expert_generator()
else: # training
current_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime())
# dir_name = datetime.now().isoformat()
args.save_dir = os.path.join('model_' + args.gan_type, current_time)
logging.info(args.save_dir)
# current_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime())
logging.debug('train {}'.format(current_time))
agent = GAN(make_env,
# load MNIST
X_train = np.load('MNIST_data/mnist_train_x.npy')
if args.model == 0:
X_train = X_train.reshape(60000, 784)
elif args.model == 1:
X_train = X_train.reshape(60000, 28, 28, 1)
X_train = tf.image.resize_images(X_train, [64, 64]).eval()
# normalize -1 to 1
X_train = (X_train.astype(np.float32) - 127.5) / 127.5
#train
model.fit(X_train)
else:
model.load(args.restoring_epoch)
if args.model == 0:
img_path = 'results/gan_genarated_image.png'
img_size = (28, 28)
elif args.model == 1:
img_path = 'results/dcgan_genarated_image.png'
img_size = (64, 64)
save_results(model,
args.restoring_epoch,
img_path,
dim=(7, 7),
figsize=(7, 7),
img_size=img_size)
