def cli(ctx, profile, snapshot):
# load hyper-parameters
hps = util.load_profile(profile)
util.manual_seed(hps.ablation.seed)
if snapshot is not None:
hps.general.warm_start = True
hps.general.pre_trained = snapshot
# build graph
builder = Builder(hps)
state = builder.build(training=False)
# load dataset
dataset = CelebA(root=hps.dataset.root,
transform=transforms.Compose(
(transforms.CenterCrop(160), transforms.Resize(128),
transforms.ToTensor())))
# start inference
inferer = Inferer(hps=hps,
graph=state['graph'],
devices=state['devices'],
data_device=state['data_device'])
ctx.obj['hps'] = hps
ctx.obj['dataset'] = dataset
ctx.obj['inferer'] = inferer
def test_glow_model(self):
# build model
hps = util.load_profile('profile/test.json')
glow_model = Glow(hps).cuda()
image_shape = hps.model.image_shape
batch_size = glow_model.h_top.shape[0]
# read image
img = Image.open('misc/test.png').convert('RGB')
x = util.pil_to_tensor(img, shape=image_shape)
x = util.make_batch(x, batch_size).cuda()
y_onehot = torch.zeros((batch_size, hps.dataset.num_classes)).cuda()
# forward and reverse flow
z, logdet, y_logits = glow_model(x=x, y_onehot=y_onehot, reverse=False)
x_ = glow_model(z=z, y_onehot=y_onehot, reverse=True)
# return parser.parse_args()
if __name__ == '__main__':
# this enables a Ctrl-C without triggering errors
signal.signal(signal.SIGINT, lambda x, y: sys.exit(0))
print("hello")
# parse arguments
#args = parse_args()
args = "profile/celeba.json"
# initialize logging
util.init_output_logging()
# load hyper-parameters
# hps = util.load_profile(args.profile)
hps = util.load_profile(args)
util.manual_seed(hps.ablation.seed)
# build graph
builder = Builder(hps)
state = builder.build()
# load dataset
dataset = CelebA(root=hps.dataset.root,
transform=transforms.Compose((
transforms.CenterCrop(160),
transforms.Resize(64),
transforms.ToTensor()
)))
# start training
def test_load_profile(self):
hps = util.load_profile('profile/celebahq_256x256_5bit.json')
self.assertIsInstance(hps, dict)