print('TYPE:', ntype)
n = Net(D, C,
type_of_net=ntype,
y_is_coded=y_coded and ntype not in ('vib', 'vae'),
batch_norm='encoder',
features='vgg16',
encoder_layer_sizes=[],
decoder_layer_sizes=[],
classifier_layer_sizes=cls_cvae if ntype == 'cvae' else [20, 10],
sigma=0,
gamma=gamma,
force_cross_y=0,
latent_sampling=L,
latent_dim=K)
n.to(d)
nets[ntype] = n
# n.compute_max_batch_size(batch_size=1024)
# print(n.max_batch_sizes)
if n.y_is_coded:
pass
if ntype != 'vae':
print('y in input')
out_y[ntype] = n.evaluate(x, y)
print('y is none')
out[ntype] = n.evaluate(x)
for o, _y in zip((out, out_y), ('*', 'y')):
jvae.saved_dir = save_dir
if args.resume:
with open(os.path.join(resumed_from, 'RESUMED'), 'w') as f:
f.write(str(job_number) + '\n')
with open(os.path.join(job_dir, f'number-{hostname}'), 'w') as f:
f.write(str(job_number + 1) + '\n')
log.info('Network built, will be saved in')
log.info(save_dir)
log.debug('%s: %s', 'Network architecture',
jvae.print_architecture(True, True))
jvae.to(device)
# print('*** .device', jvae.device)
x, y = torchdl.get_batch(trainset, device=device, batch_size=8)
if debug:
log.debug('Trying a first pass')
log.debug('x in [%.2f, %.2f] with mean (std) %.2f (%.2f)',
x.min().item(),
x.max().item(),
x.mean().item(),
x.std().item())
outs = jvae(x, y if jvae.y_is_coded else None)
log.debug(' -- '.join(map(str, ([tuple(u.shape) for u in outs]))))