def main():
# Parse config json.
cfg = process_config()
# Set the seed.
set_experiment_seed(cfg.seed)
# Create directories to be used in the experiments.
cfg = create_dirs(cfg)
if cfg.benchmark_mode:
torch.backends.cudnn.benchmark = True
# Declare instance for GAN.
if cfg.gan_type == 'WGAN':
gan = WGAN(cfg)
elif cfg.gan_type == 'LWGAN':
gan = LWGAN(cfg)
elif cfg.gan_type == 'WGAN_GP':
gan = WGAN_GP(cfg)
else:
raise Exception("[!] There is no option for " + cfg.gan_type)
# Save the hyperparameter json.
save_hparams(cfg)
# Launch the graph in a session.
gan.train()
print(" [*] Training finished!")
# Visualize learned generator.
gan.visualize_results(cfg.epoch)
print(" [*] Testing finished!")
'wait': 0
})
# Enter the training loop.
trainer.train(model,
loaders['train'],
maxepoch=config.optim.epochs,
optimizer=optimizer)
# Pick the best model according to validation score and test it.
model.reset_meters()
best_model_path = os.path.join(dirs.best_path, "best_model.pt")
if os.path.exists(best_model_path):
model.load_state_dict(torch.load(best_model_path))
if loaders['test'] is not None:
print("Testing the best model. ")
logger.log_meters('test', trainer.test(model, loaders['test']))
return model
if __name__ == '__main__':
# Get the config, initialize the model and construct the data loader.
cfg = process_config()
model_initialization = get_model(cfg)
print(model_initialization)
data_loaders = load_data(cfg)
# Train.
trained_model = train(model_initialization, data_loaders, cfg)