-
Notifications
You must be signed in to change notification settings - Fork 11
/
celeba.py
49 lines (39 loc) · 1.26 KB
/
celeba.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
import torch
from torchvision import datasets, transforms
class UnNormalize:
def __init__(self, mean, std):
self.mean = mean
self.std = std
def __call__(self, tensor):
"""
Args:
tensor (Tensor): Tensor image of size (B, C, H, W) to be normalized.
Returns:
Tensor: Normalized image.
"""
with torch.no_grad():
for i, (m, s) in enumerate(zip(self.mean, self.std)):
tensor[:, i,...].mul_(s).add_(m)
# The normalize code -> t.sub_(m).div_(s)
return tensor
def load_celeba(root, imgsz):
transform = transforms.Compose([
# transforms.RandomSizedCrop(224),
# transforms.RandomHorizontalFlip(),
transforms.Resize([imgsz, imgsz]),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
])
db = datasets.ImageFolder(root, transform=transform)
return db
def unnorm_(*args):
"""
conduct reverse normalize on each tensor in-place
:param args:
:return:
"""
net = UnNormalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
for img in args:
net(img)