alpha_affine=1,
sigma=10),
albumentations.ShiftScaleRotate(p=0.7, rotate_limit=10),
ResizeMask(h=64, w=64),
NormalizeMask(dim=(0, 1, 2))
])),
ToTensor(device),
])
R_t = DualTransformRegularizer.__call__(
g_transforms, lambda trans_dict, img: hm_svoego_roda_loss(
encoder_HG(trans_dict['image']), trans_dict['mask'], 1, 0.1))
for i in range(100000):
counter.update(i)
w300_batch = next(LazyLoader.w300().loader_train_inf)
w300_image = w300_batch['data'].to(device)
w300_mes = ProbabilityMeasureFabric(256).from_coord_tensor(
w300_batch["meta"]["keypts_normalized"]).cuda()
w300_target_hm = heatmaper.forward(w300_mes.probability,
w300_mes.coord * 63).detach()
content300 = encoder_HG(w300_image)
loss_or_none = (
writable("real_content loss", hm_svoego_roda_loss)(content300,
w300_target_hm) +
writable("R_t", R_t.__call__)(w300_image, content300) * 0.05)
pred_xy, _ = heatmap_to_measure(pred)
t_xy, _ = heatmap_to_measure(target)
return Loss(nn.BCELoss()(pred, target) +
nn.MSELoss()(pred_xy, t_xy) * 0.0005 +
(pred - target).abs().mean() * 0.3)
R_t = DualTransformRegularizer.__call__(
g_transforms, lambda trans_dict, img: hm_svoego_roda_loss(
encoder_HG(trans_dict['image']), trans_dict['mask']))
for epoch in range(130):
for i, batch in enumerate(LazyLoader.w300().loader_train):
# print(i)
counter.update(i + epoch * len(LazyLoader.w300().loader_train))
data = batch['data'].to(device)
mes = ProbabilityMeasureFabric(256).from_coord_tensor(
batch["meta"]["keypts_normalized"]).cuda()
target_hm = heatmaper.forward(mes.probability, mes.coord * 63)
content = encoder_HG(data)
hm_svoego_roda_loss(content, target_hm).minimize_step(cont_opt)
if i % 5 == 0:
real_img = next(LazyLoader.celeba().loader).to(device)
content = encoder_HG(data)
coefs = json.load(open("../parameters/content_loss_sup.json"))
R_t(real_img,
content).__mul__(coefs["R_t"]).minimize_step(cont_opt)