folder_name="AE_samples_train"),
# DiscriminatorOverfitMonitor(dataset, valid_dataset, 100, args),
ModelSaver(output_path, n=1, overwrite=True, print_output=True),
]
reconstruction_loss_mode = "pixelwise" if not args.use_dis_l_reconstruction_loss else "dis_l"
if frs_model is not None:
reconstruction_loss_mode = "frs"
train_loop = ALITrainLoop(listeners=listeners,
Gz=Gz,
Gx=Gx,
D=D,
optim_G=G_optimizer,
optim_D=D_optimizer,
dataloader=dataloader,
cuda=args.cuda,
epochs=args.epochs,
morgan_alpha=args.morgan_alpha,
d_real_label=args.d_real_label,
d_img_noise_std=args.instance_noise_std,
decrease_noise=True,
use_sigmoid=True,
reconstruction_loss_mode=reconstruction_loss_mode,
frs_model=frs_model,
r1_reg_gamma=args.r1_gamma,
non_saturating_G_loss=args.ns_gan,
disable_D_limiting=args.no_D_limit)
train_loop.train()
output_reproductions=True,
discriminator_output=True,
cuda=args.cuda,
sample_reconstructions=True,
every_n_epochs=10,
output_latent=True,
output_grad_norm=True,
ns_gan=args.ns_gan,
)
]
trainloop = ALITrainLoop(
listeners,
Gz,
Gx,
D,
G_optimizer,
D_optimizer,
dataloader,
cuda=args.cuda,
epochs=args.epochs,
morgan_alpha=args.morgan_alpha,
d_img_noise_std=args.instance_noise_std,
decrease_noise=True,
r1_reg_gamma=args.r1_gamma,
non_saturating_G_loss=args.ns_gan,
disable_D_limiting=args.no_D_limit
)
trainloop.train()